Three-Dimensional Three-Component Air Flow Visualization in a Steady-State Engine Flow Bench Using a Plenoptic Camera

Plenoptic particle tracking velocimetry (PTV) shows great potential for three-dimensional, three-component (3D3C) flow measurement with a simple single-camera setup. It is therefore especially promising for applications in systems with limited optical access, such as internal combustion engines. The 3D visualization of a plenoptic imaging system is achieved by inserting a micro-lens array directly anterior to the camera sensor. The depth is calculated from reconstruction of the resulting multi-angle view sub-images. With the present study, we demonstrate the application of a plenoptic system for 3D3C PTV measurement of engine-like air flow in a steady-state engine flow bench. This system consists of a plenoptic camera and a dual-cavity pulsed laser. The accuracy of the plenoptic PTV system was assessed using a dot target moved by a known displacement between two PTV frames. In this manner, the PTV accuracy at different seeding densities, particle displacements for lateral and depth directions, and different depth planes was evaluated. Then, 3D3C PTV was conducted on a steady-state engine flow bench at two different flow conditions; one with strong tumble, the other with a combination of tumble and swirl. Silicone oil droplets were seeded into the air flows and were illuminated by the volume-expanded laser beam. Mie scattering from the droplets was recorded by the plenoptic camera running in double-frame mode. A sequence of double-frame images was required to produce an averaged 3D velocity field. The 3D3C velocity fields for both flow conditions are clearly captured, demonstrating the plenoptic 3D PTV technique as a feasible tool to investigate the complex 3D engine flow.


  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01638472
  • Record Type: Publication
  • Source Agency: SAE International
  • Report/Paper Numbers: 2017-01-0614
  • Files: TRIS, SAE
  • Created Date: Jun 23 2017 11:40AM